Prediction of Occluded Pedestrians in Road Scenes using Human-like Reasoning: Insights from the OccluRoads Dataset

TL;DR

This paper introduces the OccluRoads dataset for occluded pedestrian detection and proposes a reasoning pipeline that significantly improves prediction accuracy by mimicking human perception in complex road scenes.

Contribution

The work presents a new dataset with diverse occlusion scenarios and a novel reasoning pipeline combining Knowledge Graphs and Bayesian inference for pedestrian prediction.

Findings

Achieved an F1 score of 0.91 in occluded pedestrian prediction.

Improved prediction performance by up to 42% over traditional models.

Provided insights into human-like reasoning for complex perception tasks.

Abstract

Pedestrian detection is a critical task in autonomous driving, aimed at enhancing safety and reducing risks on the road. Over recent years, significant advancements have been made in improving detection performance. However, these achievements still fall short of human perception, particularly in cases involving occluded pedestrians, especially entirely invisible ones. In this work, we present the Occlusion-Rich Road Scenes with Pedestrians (OccluRoads) dataset, which features a diverse collection of road scenes with partially and fully occluded pedestrians in both real and virtual environments. All scenes are meticulously labeled and enriched with contextual information that encapsulates human perception in such scenarios. Using this dataset, we developed a pipeline to predict the presence of occluded pedestrians, leveraging Knowledge Graph (KG), Knowledge Graph Embedding (KGE), and a Bayesian inference process. Our approach achieves a F1 score of 0.91, representing an improvement of up to 42% compared to traditional machine learning models.